Hybrid molecules as strategy for the design of new anti-infective agents International Symposium

1
Hybrid molecules as strategy for the designof
new anti-infective agents International
Symposium  From synthetic chemistry to synthetic
biology Collège de France Paris, 5 mai 2009

• Bernard MEUNIER

Laboratoire de Chimie de Coordination du CNRS,
Toulouse (1979-2006) et PALUMED,
Toulouse-Castanet (depuis février 2006)
2
A hard time for Drug Discovery !

• Drug discovery is highly challenging
• Difficulties to create new drugs
• Economical constraints (new drugs should be
  cheap, even at no-profit level in some cases)
• Longer time  from Patent to Market  12-15
  years, compared to 8 years in the 1960s.
• Increase of the RD costs 0.8 to 1.4 billion
  USD !
• Break of costs 10 for discovery, 15 for
  pre-clinical, 15 for manufacturing and process,
  55 for clinical trials and 5 for marketing.

3
Decrease of the number of approved drugs
Only 17 new molecular entities approved by the
FDA in 2007.
Molecules
Costs
4
Genomics as an answer for drug discovery ?

• - Glaxo (GSK) spent 7 years on genomic studies on
  bacteria to evaluate more than 300 gene products
  as potential targets.
• 70 high-through-put screening (HTS) were then
  performed without success (with large libraries
  300,000 to 500,000 chemicals).
• Total cost of the GSK campaign was above 150 M
  .
• Conclusion there is no  fast-track  from gene
  to target.

5
Future trends in drug discovery

• - Back to natural products as source of new drugs
  and as inspiration for new scaffolds ! (We should
  remember Pierre Potiers predictions).
• Chemical genetics study of a gene function
  with chemical tools (not by genetic knockout as
  in classical genetics).
• Observations during clinical trials are in
  fact reverse chemical genetics ! The phenotypic
  effect of Viagra on the erectile function has
  been discovered during a clinical trial as
  vasodilatator for the treatment of heart disease.
• - Dynamic combinatorial librairies (Lehn et al.)
• Development of computational methods to perform
  virtual screening of targets.

6
Small molecules as drugs the downfall or a real
future ?

• - Fast development of biopharmaceuticals
  (proteins, DNA or RNA) over the last two decades.
• New efficient vaccines.
• Highly specific antitumoral antibodies.etc.
• In 2008, the drug market is still dominated by
  small molecules (80), but this ratio will
  decrease, slowly or quickly ? (the high cost of
  biological drugs will probably limit their
  developments).

7
Limits for health costs ?

• - Health costs reached 16 of the gross domestic
  product (GDP) in the USA in 2005.
• What will the limit (or the plateau) for health
  costs in a developed country 20, 25 ?
• A Swedish survey in 2003 indicates that the
  annual cost of antirheumatic antibodies is 12,000
  /year, compared to a cost of 170 /year/person
  for chemical drugs. 12,000 euros the price of a
  medium-size car.

8
Did we reached the limits of the vast chemical
space ? No
Fink and Reymond generated in 2007 a database of
26 million molecules with up to 11 atoms of
carbon, nitrogen, oxygen and fluorine that are
feasible (only 63,850 molecules of this limited
chemical space are already in public databases,
i.e. 0.24).
Cartoon view of the chemical space and discrete
areas occupied by inhibitors of kinases,
proteases, .etc.
9
Why making hybrid molecules as potential drugs ?

• - Nature is making hybrid molecules !
• The antitumoral bleomycin is a good example of a
  hybrid molecule with three different domains (DNA
  binding, metal binding and cell penetration).

10
Different strategies for making hybrid molecules
A one single target (double-edged sword
molecules)
single target
single target
B two independent targets (the two entities of
the hybrid molecule act independently on the
targets)
target 2
target 1
C two related targets (both entities of the
hybrid molecule act at the same time)
target 2
target 1
B. Meunier, Acc. Chem. Res., 41, 69-77 (2008)
11
Mechanism-driven design of trioxaquines

• Requirements for antimalarial drug design
• bitherapy is recommended to avoid the emergence
  of resistant parasite strains (gt hybrid
  molecules)
• (ii) different pharmacokinetics of two
  independent drugs are generating
• difficulties in bitherapy.

Trioxaquines are based on a strategy using
hybrid molecules with a dual mode of action.
The two active entities are covalently linked to
synchronize their biodisponibility.
Trioxane-linker-quinoline  trioxaquine 
The trioxaquines are highly active on
chloroquine-resistant Plasmodium falciparum.

Dechy-Cabaret et al., CNRS patent, April 2000.  
For a review, see Meunier, Acc. Chem. Res.,
2008, 69-77
12
Current status of the development of trioxaquines

• 120 trioxaquines and trioxolaquines have been
  prepared
• by Palumed and evaluated in vitro (IC50 values)
  between
• February 2003 and December 2006.
• 77 of these hybrid molecules have been evaluated
  in vivo
• (mice model, determination of CD50 and CD90
  values).
• 6 of these trioxaquines have been further
  evaluated.
• PA1103 has been selected in January 2007 for
• development by Sanofi-aventis ( SAR116242).
• - 3 kg have been prepared in March 2008. 12 kg of
  the GMP product are currently in production.

13
Structure of the trioxaquine PA1103/SAR116242

• PA1103 is a 50/50 mixture of two diastereoisomers

Coslédan et al., PNAS, vol. 105, 17579-17584
(2008)
14
Properties of PA1103 (selected as drug-candidate)

• Molecular weight 460 (base form) (OK with the
  Lipinskis rules)
• Crystalline form, log P calc. 4.9
• Stable at 50 C for months.
• Stable at acid pH values for hours.
• Ames negative.
• Cardiosafety is OK.
• Good metabolic profile (only one main
  metabolite).
• PK studies (rat) biodisponibility 30-35.
• IC50 value on P. falciparum CQ-resistant FcM29
  9 nM.
• IC50 values on six different strains from 7 to
  15 nM.
• PA1103 is active on clinical isolates (Africa).

15

• PA1103 has a dual mode of action
• . Inhibition of heme polymerization like
  chloroquine.
• . Alkylation of heme like artemisinin.
• . Active on the early stages of the
  multiplication within the red bloods
  like artemisinin.
• . Active on gametocytes like artemisinin.

Coslédan et al., PNAS, vol. 105, 17579-17584
(2008).
16
New antibiotics a real medical need

•  After the decline facilitating a renewal in
  antibiotic development  WHO (2004) and EASAC
  (European Academies Science Advisory Council,
  2007).
•  Despite the critical need for new antimicrobial
  agents, the development of these agents is
  declining. Solutions encouraging and facilitating
  the development of new antimicrobial agents are
  needed  (Spellberg et al., Clinical Infectious
  Diseases, 2004, 38, 1279-1286).
• In 2004, among 506 drug-candidates in development
  in pharma and biotech companies, 67 are for
  cancer, 33 for inflammation, 34 for metabolic
  disorders and only 6 for bacterial infectious
  diseases (same authors as above).

17
PALUMED has identified three different targets
for new antibiotics

• Nosocomial infections (vancomyquine).
• The number of MRSA-related hospitalization
  doubled within 7 years (1999-2005) and the number
  of deaths due to nosocomial infections
  (19,000/year) is higher than that related to AIDS
  (E. Klein et al. CDC-Atlanta, www.cdc.gov.eid,
  December 2007).
• An anti-MRSA antibiotic active by oral
  administration (cephaloquine).
• A cephalosporin-antibiotic active by oral route
  is deeply needed.
• An antibiotic active on drug-resistant
  gram-negative bacteria (bactamiquine).
• Multidrug-resistant gram-negative bacteria are
  responsible for nosocomial pneumonia-attributed
  mortality (Pseudomonas aeruginosa, Acinetobacter
  spp, ).
• Names in blue are
  corresponding to PALUMEDs programs.

18
Antibioquines as new antibiotics

• - To fight bacteria resistant to classical
  antibiotics, PALUMED extended the concept of
  hybrid molecules named  antibioquines  (Patent
  applications 2004, 2009).
• Concept covalent attachment of an
  aminoquinoline entity (AQ) to an antibiotic
  skeleton.

19
Bactericidal activities of vancomyquines compared
to vancomycin, linezolid, telavancin and
daptomycin against S. aureus MRSA with 50 of
human serum
Vancomycin
Telavancin
Daptomycin
PA1409
PA1410
PA1247
PA1274
PA1418
Q L N-R MIC without serum MIC with serum
- - - 1 2
- - - 0.5 4
- - - 0.125 0.5
- - - 0.5 0.5
- - - 0.5 2
- - 0.125 0.5
- - - 0.25 0.5
- - - 0.25 0.25
Concentration 1 mg/mL
(MIC values obtained by macro-methods)
Control
Vancomycin
Only vancomyquines are able to reduce the
bacterial colonies by 4.5 log units within 24 h
at 1 mg/mL compared to vancomycin,
telavancin, linezolide or daptomycin.
Daptomycin
Linezolid
D log cfu
Telavancin
PA1410
PA1247 PA1274 PA1409 PA1418
For PA1409 1 ?g/mL 0.57 ?M
Time (h)
20
PA1409 influence of inoculum on its bactericidal
activity against S. aureus MRSA (clinical isolate
mR) with 50 of human serum
MICs (values obtained by macro-method,
bactericidal conditions)
PA1409
Vancomycin
Telavancin
Daptomycin
Inoculum 105 CFU/mL
1 2
2 2
0.125 0.5-1
0.5 2
MIC without serum MIC with serum
PA1409 is the most active at a low dose
Inoculum 107 CFU/mL
2 2
16 8
0.5 1
2 2
MIC without serum MIC with serum
Bactericidal activities at a concentration 2
mg/mL
Small inoculum 105 CFU/mL
High inoculum 107 CFU/mL
Control
Control
D log cfu
Vancomycin
PA1409
Telavancin
Daptomycin
Vancomycin
Telavancin
Daptomycin
PA1409
Time (h)
Time (h)
21
Bactericidal activity on E.faecalis VSE (Isolat
U38) or VRE Van A (CIP106996) with 50 of human
serum at 4 µg/mL
Studies 08-831/37 et 07-831/23
Control
Control
Vancomycin
Telavancin
Vancomycin
Daptomycin
Telavancin
PA1409
Daptomycin
PA1409
22
Bactericidal activity on E.faecium VSE (Isolat
B1003) or VRE Van A (CIP107387) with 50 of human
serum at 16 µg/mL
Studies 08-831/38 et 08-831/29M
Control
Control
Vancomycin
Telavancin
PA1247
Vancomycin
PA1409
PA1409
PA1247
Telavancin
23
Vancomyquines highly active in vivo on MRSA and
PRSP (mice)

• - Vancomyquine PA1409 is curative by iv route on
  mice infected by MRSA (septicemia) CD100 5
  mg/kg.
• The CD50 value of PA1409 on MRSA 1 mg/kg.
• (infection by sc, iv treatment at t 1 h and
  t 4 h)
• On MSSA CD80 of PA1409 4 mg/kg.
• On the same MRSA model, the CD100 values of
  competitors are
• Vancomycin CD100 20 mg/kg (no cured mice at 1
  mg/kg, CD50 gt 5 mg/kg).
• Telavancin CD100 above 20 mg/kg (CD50 10
  mg/kg).
• Daptomycin CD100 10 mg/kg (CD50 6 mg/kg).
• PA1409 is more potent than all the other
  competitors.
• Vancomyquines are curative at 3 mg/kg (sc route)
  on mice infected by PRSP (no cured animals with
  vancomycin in the same conditions).

24
AUC/MIC ratios of PA1409 and vancomycin
Dog - adult beagle - 2.5 mg/kg - iv injection -
glucose 5 - Win Non Lin version 5.2 software
analysis
PA1409 Vancomycin
AUC (minxmg/ml) 17105 755 1143 40
MIC90 (mg/ml) 0.25 1
AUC/MIC 68420 1143
MIC90 from eleven MRSA strains
The AUC/MIC ratio for PA 1409 is 60 times that of
vancomycin.
25
Acknowledgements
Mechanism of action of artemisinin
derivatives Anne ROBERT (CNRS Fellow), Jérôme
CAZELLES (PhD 2000), Monserrat
RODRIGUEZ (post-doct, Spain) Katalina SELMECZI
(post-doc, Hungary), Sophie A. -L.
LAURENT (PhD 2006) Fatima BOUSEJRA-EL GARAH (PhD
student) Financial support CNRS, PALUMED, ANR
and EU-Antimal Synthesis of trioxaquines Frédér
ic COSLEDAN (Palumed), Christine SALLE
(Palumed), Odile DECHY-CABARET (PhD 2001),
Christophe LOUP (CNRS), Jérôme CAZELLES
(Palumed), Anne ROBERT (CNRS Fellow) Heinz
GORNITZKA (X-ray structures) Financial support
PALUMED, CNRS, ANR, Conseil Régional
Midi-Pyrénées and EU-Antimal Biological
activities of trioxaquines Frédéric COSLEDAN
(Palumed), Françoise BENOIT-VICAL (INSERM
Fellow), Joël LELIEVRE (PhD), Angélique ERRAUD
(Palumed), Carine AUGE (Palumed), Céline BERRONE
(Palumed), Katia JONOT (Palumed) Support
PALUMED, ANR, CHU-Rangueil and EU-Antimal
Academic collaborations J. F. MAGNAVAL, J. P.
SEGUELA and A. BERRY (Toulouse Hospital,
CHU-Rangueil), P. KREMSNER (Lambaréné, Gabon), D.
DIVES (Lille, Inserm), D. MAZIER (Paris,
Inserm). Collaboration with Sanofi-Aventis
Jean-Pierre MAFFRAND, Laurent FRAISSE, Alain
PELLET For informations on
PALUMED see www.palumed.com
26
Acknowledgements (antibiotics)
Synthesis and PK data of antibioquines Muriel
SANCHEZ, Jérôme CAZELLES, Michel NGUYEN,
Camille CATHARY, Laurence PAGES Biological
activities of antibioquines Charlotte
DUVAL Collaborations Christine ROQUES,
Faculty of Pharmacy - Toulouse (FONDEREPHAR) Pier
re-Louis TOUTAIN, Veterinary School of
Toulouse Roland LECLERCQ (CHU-Caen) Bruno
FANTIN (CHU-Beaujon, Paris) Financial support
PALUMED